Summary There are striking population (race) disparities in prostate cancer (PCa) risk and survival outcome borne out of current health statistics. This is particularly evident between African American (AA) patients and their European American (EA) counterparts, where AAs exhibit a 1.5 to 2 fold higher risk of PCa incidence and mortality. We demonstrate that differential alternative RNA splicing takes place for oncogenes and tumor suppressor genes in PCa specimens of AA compared to EA patients. The differential splicing events result in the enrichment and in many instances generation of AA-specific splice variants not observed in EA cancers. We have cloned the AA and EA splice variant cDNAs for both phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit delta (PIK3CD) and fibroblast growth factor receptor 3 (FGFR3), and ectopically introduced the clones individually into PCa cell lines. Compared to the EA variant-expressing lines, the AA variant-expressing lines are more oncogenic in in vitro cell proliferation and invasion assays, and exhibit greater tumor growth/metastasis and resistance to small molecule targeted therapy in xenograft mouse models. These findings may explain, in part, the observed cancer health disparities in the AA population relative to other racial groups. The mechanism responsible for the formation of AA-specific or -enriched mRNA splice variants are hypothesized to be due to the dysregulated expression of 6 key splicing factor s in AA PCa specimens. We refer to this dysregulation and associated production of AA-specific splice variants as an ?AA splicing factor code in PCa disparities.? Notwithstanding these developments, a number of questions remain unresolved. First, what is the mechanism(s) of dysregulated expression of the 6 splicing factors? Second, can the production of AA-specific splice variants via exon skipping be directly attributed to the 6 dysregulated splicing factors? Lastly, how do the encoded proteins of the AA-specific short variants of PIK3CD and FGFR3 mediate a more oncogenic phenotype in PCa cells? The objectives of this application are to address these questions, and to provide a mechanistic framework of population-related differences in alternative RNA splicing, which is anticipated to facilitate future identification of druggable targets (e.g. splicing factors and/or resulting splice variant proteins) of AA tumor phenotype and disparities.